On the other hand, data (and data efficiency) is still a problem. This is not the sort of thing that gets us beyond a sim-to-real training paradigm, so it might be used but it’s not a big deal.
On the other hand, data (and data efficiency) is still a problem. This is not the sort of thing that gets us beyond a sim-to-real training paradigm, so it might be used but it’s not a big deal.